# Scaling Graph Neural Networks for Friend Recommendation: Multi-Hash User Embeddings and Temporal Neighbor Sampling

arXiv: https://arxiv.org/abs/2608.27413 · HTML: https://mlsanity.com/paper/2608.27413/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 28 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation](https://mlsanity.com/paper/2002.02126/) (2020)
- [Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https://mlsanity.com/paper/1806.01973/) (2018)
- [How Attentive are Graph Attention Networks?](https://mlsanity.com/paper/2105.14491/) (2021)
- [Neural Graph Collaborative Filtering](https://mlsanity.com/paper/1905.08108/) (2019)
- [Temporal Graph Networks for Deep Learning on Dynamic Graphs](https://mlsanity.com/paper/2006.10637/) (2020)
- [Inductive Representation Learning on Temporal Graphs](https://mlsanity.com/paper/2002.07962/) (2020)
- [Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems](https://mlsanity.com/paper/1909.02107/) (2019)
- [TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphs](https://mlsanity.com/paper/2203.14883/) (2022)
- [Hash Embeddings for Efficient Word Representations](https://mlsanity.com/paper/1709.03933/) (2017)
- [Learning to Embed Categorical Features without Embedding Tables for Recommendation](https://mlsanity.com/paper/2010.10784/) (2020)
- [LiGNN: Graph Neural Networks at LinkedIn](https://mlsanity.com/paper/2402.11139/) (2024)
- [GiGL: Large-Scale Graph Neural Networks at Snapchat](https://mlsanity.com/paper/2502.15054/) (2025)
- [GraphStorm: all-in-one graph machine learning framework for industry applications](https://mlsanity.com/paper/2406.06022/) (2024)
- [Position-based Hash Embeddings For Scaling Graph Neural Networks](https://mlsanity.com/paper/2109.00101/) (2021)
- [Ultra Fast Warm Start Solution for Graph Recommendations](https://mlsanity.com/paper/2509.01549/) (2025)
- [GNN Applied to Ego-nets for Friend Suggestions](https://mlsanity.com/paper/2412.11888/) (2024)

## Abstract (arXiv)

Friend recommendation is inherently graph-structured: the relevance of a potential connection depends on multi-hop social context rather than user attributes alone. However, deploying message-passing GNNs on a production-scale social graph with hundreds of millions of users and tens of billions of edges requires addressing numerous modeling and systems challenges. We present a scalable end-to-end GNN ranking system for production social graphs, focusing on two design choices that are critical in this setting: multi-hash ID embeddings and temporal neighbor sampling. Multi-hash embeddings are common for high-cardinality features, but industrial GNN systems typically either ignore trainable IDs or accept full embedding tables, exceeding 200 GB for our graph. We integrate multi-hash as the primary node representation, reducing the ID-embedding table size by more than 98 percent while preserving ranking quality. Temporal neighbor sampling is well understood in principle, but existing implementations scan full adjacency lists, which is a non-starter for users with tens of thousands of friends. We implement timestamp-sorted CSR storage with binary search, reducing the per-node temporal sampling cost from O(deg(v) + k) to O(log(deg(v)) + k). Beyond these components, we show that this combination scales and yields measurable production impact. On a graph with 194M users and 28B edges, offline ablations isolate each design choice's contribution. In an online A/B test, our system increases friend additions from recommendations by 16 percent and unique friend adders by 11.5 percent over a strong production baseline. We release our framework for distributed training and inference on large temporal graphs.